The 3rd Generation Partnership Project, 3GPP, is responsible for the standardization of the Universal Mobile Telecommunication System (UMTS) and Long Term Evolution, LTE. The 3GPP work on LTE is also referred to as Evolved Universal Terrestrial Access Network (E-UTRAN). LTE is a technology for realizing high-speed packet-based communication that can reach high data rates both in the downlink and in the uplink, and is thought of as a next generation mobile communication system relative to UMTS. In order to support high data rates, LTE allows for a system bandwidth of 20 MHz, or up to 100 MHz when carrier aggregation is employed. LTE is also able to operate in different frequency bands and can operate in at least Frequency Division Duplex (FDD) and Time Division Duplex (TDD) modes.
Hybrid Automatic Repeat reQuest (HARQ) is an integral part of the 3G and 4G standards that allows reliable communication between a wireless device and a network node by means of incremental redundancy. The transport block to be transmitted is subjected to forward error correction encoding by which redundancy is introduced. The number of bits increases due to the introduced redundancy, but not all bits are sent at the same time. The resulting bits are segmented into several so called redundancy versions, where each such redundancy version further is punctured before being sent in order to fit it within the given allocation (one or more resource block pairs). How much is punctured is depending on how many bits (information plus redundant bits) that can be carried in the allocation, which further is depending on the allocation bandwidth, the modulation (e.g. QPSK, 16QAM, 256QAM) in use, and the presence of broadcasted signals and channels in the allocated bandwidth. The ratio between the information bits and information bits plus redundant bits in a transport block is referred to as code rate. The combination of code rate and modulation type is referred to as Modulation and Coding Scheme.
In case the maximum number of retransmissions is reached without the receiving entity being able to decode the transport block, it will be detected by higher layers e.g. Radio Link Control, RLC, generally within 50-100 ms that a Protocol Data Unit (PDU) is missing and a retransmission is requested for all transport blocks that comprise the RLC PDU, even those that may have been successfully. This is referred to as Automatic Repeat request (ARQ) and has considerably larger latency than HARQ retransmissions.
In a LTE network, a wireless device (in LTE referred to as a User Equipment, UE) carries out measurements to provide indications to the base station (in LTE eNodeB) on the perceived radio propagation conditions in what is called Channel Quality Indicator (CQI) reporting. Based on the reporting the base station can decide roughly which Modulation and Coding Scheme (MCS) to use for communication with the UE. One example of mapping between CQI and MCS is shown in Table 1 below, which is the retrieved from 3GPP TS 36.213 V10.12.0 section 7.2.3. In low channel quality (low CQI index) more forward error correction encoding is needed for successful decoding of the information bits, and vice versa in high channel quality i.e. high CQI index. Hence at high CQI the throughput of information bits can be made higher than at low CQI.
TABLE 14-bit CQI table from 3GPPCQIcode rate ×indexmodulation1024efficiency0out of range1QPSK780.15232QPSK1200.23443QPSK1930.37704QPSK3080.60165QPSK4490.87706QPSK6021.1758716QAM3781.4766816QAM4901.9141916QAM6162.40631064QAM4662.73051164QAM5673.32231264QAM6663.90231364QAM7724.52341464QAM8735.11521564QAM9485.5547
In order to get a good system throughout the base station, in LTE referred to as an eNB, carries out link adaptation by which it matches each UE's reported channel quality to an MCS that provides the right balance between system throughput and throughput for the individual user. The MCS is indicated to the UE in the Downlink Control Information (DCI) provided over Physical Data Control Channel, PDCCH in LTE. This is shown in Table 2 which is the retrieved from 3GPP TS 36.213 V10.12.0 section 7.1.7.1.
In addition to CQI reporting, a base station typically has an outer loop that based on ACK/NACK reports tunes in the MCS value to a suitable value giving a BLER (ratio between NACKs and total number of received or expected ACK/NACKs) of e.g. 10%. Besides catering for flexibility in which target BLER is used (e.g. 1%, 10%, 30%), it also solves the problem that each UE model or even UEs of the same model may have an individual bias in the reported CQI. The base station thus maintains a UE-specific CQI offset which it tunes to give the desired BLER target.
TABLE 2Modulation and TBS index table for PDSCHMCSModulationTBSIndexOrderIndexIMCSQmITBS0201212223234245256267278289291049114101241113412144131541416415176151861619617206182161922620236212462225623266242762528626292reserved304316
Existing control loops on base station side for determining the UE-specific MCS are largely based on maintaining a particular BLER ratio and identifying UE-specific bias in the CQI reporting. Once the CQI bias has been identified the base station can more accurately select MCS based on the CQI reported by the UE. During identification of the CQI bias the base station will decrease or increase the MCS until the BLER target is met over some interval of time.
The expansion of machine-type communication (MTC) toward industrial applications is seen as one of the key features in future communication systems. The requirements on connectivity are very diverse and largely depend on the use case of an industrial application to be operated. Therefore, different Critical-MTC (i.e., ultra-reliable MTC) solutions will be needed. Besides the end-to-end latency, the Critical-MTC concept should address the design trade-offs regarding transmission reliability, mobility, energy-efficiency, system capacity and deployment, and provide solutions for how to design a wireless network in a resource and energy efficient way while enabling ultra-reliable communication.
For scheduling of UEs that are to fulfill extreme requirements on BLER, e.g. down to 10−9, while at the same time fulfilling extreme requirements on latency, i.e. being able to send and receive new information every sub frame, block errors have to be avoided as far as possible.
With existing implementation it may however be problematic to adjust to e.g. a proper MCS without introducing block errors occasionally and as a result the base station (or network node) is forced to be very conservative in the MCS selection. Being conservative implies using a lower/less aggressive MCS than called for, resulting in that more resources are used for the particular UE than necessary, with reduced system throughput as result.